© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 39812
    Prev
    Next
    INDEX
    Explanations

    colourful description

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    å®īä¿Ŀ
    -0.10
    åĹŁ
    -0.10
     Scotia
    -0.10
    è¨Ĺ
    -0.10
    çĻ½èī²çļĦ
    -0.10
    俵
    -0.10
    å®£ä¼łçīĩ
    -0.09
    çĻ½èī²
    -0.09
    æŀĩ
    -0.09
    æµģ浪
    -0.09
    POSITIVE LOGITS
    æĸijæĸĵ
    0.11
    é²ľèī³
    0.09
    ãĤ«ãĥ©
    0.08
    æĹĹå¸ľ
    0.08
    ful
    0.08
    oko
    0.08
    æ¾
    0.08
    对åŃ©åŃIJ
    0.08
    迦
    0.08
    OTE
    0.08
    Activations Density 0.095%

    No Known Activations